Abstract
Objectives
To investigate whether poorer survival of breast cancer among deprived women compared with affluent women is related to their NHS care.
Design
Retrospective review of hospital and general practice case records.
Setting
Greater Glasgow Health Board area.
Subjects
Women diagnosed with breast cancer in 1992-3 who lived in the most affluent (deprivation categories 1 and 2) and the most deprived areas (deprivation categories 6 and 7) of Glasgow (Carstairs and Morris deprivation index).
Main outcome measures
Breast cancer treatment, time from general practice consultation to clinic visit and surgery, and details of hospital admissions and follow up in primary and secondary care.
Results
The access to care and surgical and oncological treatment of women from affluent and deprived areas were similar. Admissions to hospital for problems not related to breast cancer were more common in those living in deprived areas (number admitted once or more: 51 (24%) v 13 (10%), P=0.001). Consultation patterns in general practice by the second year after diagnosis showed women in deprived areas consulting more frequently than women in affluent areas (median (interquartile range) number of consultations (5 (2-10) v 7 (4-13), P=0.01).
Conclusion
Women living in affluent areas did not receive better NHS care for breast cancer than women in deprived areas. However, women from deprived areas seem to have greater comorbidity, and poorer outcomes from breast cancer among these women is probably due to factors which result in deprived communities having poorer health outcomes rather than to management of their breast cancer.
Introduction
Affluent women have a higher incidence of breast cancer than socially deprived women.1 Several studies have shown deprived women to have poorer survival from breast cancer.2–6 Two of these studies also investigated the relation of socioeconomic status to pathological prognostic factors at presentation and found no association.5,6 Our study of pathological prognostic factors at presentation confirmed these findings for women with early breast cancer.7 If pathological staging is not associated with the poorer outcome of deprived women, the question arises whether poorer outcome is the result of differing care and treatment of breast cancer among women from deprived and affluent areas.
Previous work has shown differences in the management of women with breast cancer—for example, between teaching and non-teaching hospitals8,9 and between specialists and non-specialists.10–12 We investigated differences in management between women of differing socioeconomic status by detailed analysis of hospital and general practice records.
Subjects and methods
The West of Scotland Cancer Surveillance Unit identified all women resident in the Greater Glasgow Health Board area in whom invasive breast cancer was diagnosed in 1992 (447 cases) and 1993 (374 cases). From this population, women were identified who lived in areas at either end of the deprivation spectrum and who were under 75 years at time of diagnosis. The Carstairs and Morris area based deprivation index was used to define deprivation (affluent: deprivation categories 1 and 2; deprived: deprivation categories 6 and 7).13 Ethical approval was obtained from the Greater Glasgow Health Board local medical ethics committee.
In order to document fully the treatment received from the NHS by women with breast cancer, we collected data from both hospital and general practice records. Hospital records were reviewed in five hospitals to obtain information about treatment received, including breast surgery (conservation surgery or mastectomy), axillary surgery (sampling or clearance, as defined by the surgeon), and whether the women received adjuvant radiotherapy, chemotherapy, or endocrine therapy. Access to care was investigated by examining delays in the process of care after presentation, length of inpatient stays, and number and nature of outpatient visits.
We identified general practitioners of individual patients from hospital case records and contacted practices to ask permission to review case records. Data collected included details at presentation and consultations with general practitioner in the 12 months before diagnosis and in the first and second 12 months after diagnosis.
Analysis
We analysed the data using SPSS software. Associations between treatment options and whether women lived in affluent or deprived areas were examined by χ2 tests. The relations of time to treatment and consultation frequency to deprivation were analysed by Mann-Whitney tests.
Results
In 1992 and 1993, 421 women were diagnosed with breast cancer in the deprivation categories being studied. Of these, 158 (38%) lived in deprivation categories 1 and 2 (affluent) and 263 (62%) in deprivation categories 6 and 7 (deprived); 417 hospital records were examined (99%). We present here data for women who had early breast cancer (146 women from affluent areas and 220 women from deprived areas). The general practitioner was known for 327 women in 138 practices; 278 (86%) of these records were examined (table 1).
Table 1.
Numbers of hospital and general practice records examined of women with breast cancer diagnosed in 1992 and 1993 and living in affluent or deprived areas
| Affluent (n=158) | Deprived (n=263) | Total (n=421) | |
|---|---|---|---|
| Hospital: | |||
| Records requested | 158 | 263 | 421 |
| Records seen | 157 | 260 | 417 |
| Diagnosis of primary breast cancer | 146 | 220 | 366 |
| Diagnosis of locally advanced or metastatic breast cancer | 10* | 40 | 50 |
| General practice: | |||
| Records requested† | 134 | 193 | 327 |
| Records seen | 110 (82%) | 168 (87%) | 278 (85%) |
One woman presented with breast cancer and ovarian cancer simultaneously and is omitted from table.
39 records were not requested because the women had moved out of the area or their general practitioner was unknown.
Access to care
Time between the date of the general practice consultation and the date on the referral letter did not differ between women living in affluent and deprived areas (affluent: median 1 day, interquartile range 0 to 2.7; deprived: median 0 day, 0 to 4; Z=−0.34, P=0.74). Only four women received private care. The time between the date of the referral letter and the first clinic visit was one day shorter in women from affluent areas (table 2). There was, however, no significant difference between the groups in the time from clinic visit to surgery or in number of days spent in hospital at the time of original surgery.
Table 2.
Access to care for breast cancer as measured by number of days from general practitioner's referral letter to clinic visit, from first clinic visit to surgery, and inpatient stay at time of initial surgery for women living in affluent and deprived areas
| Affluent
|
Deprived
|
Mann-Whitney test | ||||
|---|---|---|---|---|---|---|
| No of women | Median (interquartile range) delay (days) | No of women | Median (interquartile range) delay (days) | |||
| Time from referral letter to clinic visit* | 117 | 6 (1-13) | 183 | 7 (4-20) | Z=−2.72 P=0.006 | |
| Time between first clinic visit and surgery* | 129 | 16 (9-24) | 203 | 17 (10-25) | Z=−1.53 P=0.13 | |
| Length of hospital stay at time of initial surgery | 142 | 7 (5-8) | 215 | 6 (5-8) | Z=−1.04 P=0.29 | |
Data presented only for women with available referral letter.
Treatment
There were no differences in the type of breast surgery between women living in affluent and deprived areas (table 3). However, more women living in deprived areas had axillary sampling rather than axillary clearance. No significant differences were detected between women living in affluent and deprived areas with respect to radiotherapy, chemotherapy, or endocrine therapy (table 3).
Table 3.
Surgical treatment, radiotherapy, and adjuvant therapy received by women living in affluent and deprived areas
| No (%) in affluent area (n=146) | No (%) in deprived areas (n=220) | χ2 test result | |
|---|---|---|---|
| Breast surgery*: | |||
| Mastectomy | 64 (45) | 104 (48) | χ2=0.37, df=1 P=0.54 |
| Conservation | 78 (55) | 111 (52) | |
| Axilla surgery†: | |||
| Clearance | 123 (95) | 146 (74) | χ2=23.73, df=1 P<0.0001 |
| Sampling | 6 (5) | 50 (26) | |
| Radiotherapy | 54 (37) | 90 (41) | χ2=0.56, df=1 P=0.45 |
| Chemotherapy | 29 (20) | 30 (14%) | χ2=2.51, df=1 P=0.11 |
| Endocrine therapy | 128 (88) | 196 (89) | χ2=0.17, df=1 P=0.67 |
3 women had no surgery and in 6 cases the procedure was unclear from the hospital records.
37 women had no axillary surgery and in 4 cases the procedure was unclear.
Follow up
The number of hospital clinic visits was not significantly different between women living in affluent and deprived areas (mean (SD) number of visits 7.63 (2.76) for women from affluent areas and 7.98 (3.14) for women from deprived areas, t=1.10, P=0.27) in the first two years after diagnosis. The proportion of women requiring extra clinic appointments before the scheduled appointments did not differ between the groups. Although the numbers are small, significantly more patients from deprived areas failed to attend appointments (6 (4%) v 25 (12%), χ2 =5.65, P=0.017).
Most women had consulted their general practitioner in the year before diagnosis, but there was no difference between those living in affluent and deprived areas (92 (86%) v 141 (88%), χ2=0.26, P=0.61). In the 12 months after diagnosis, all women consulted more often than before diagnosis, but women living in deprived areas consulted more frequently than women living in affluent areas (table 4). The frequency of consultations in the second year after diagnosis was still greater than before diagnosis, with women in deprived areas continuing to consult more often than women in affluent areas. This difference was particularly striking in women who consulted more than once a month. During the first year after diagnosis, 31 (29%) women living in affluent areas consulted their general practitioner more than once a month, compared with 67 (41%) women living in deprived areas (χ2 =4.49, df=1, P=0.034). In the second year after diagnosis, 17 (16%) women living in affluent areas were continuing to consult more frequently than once a month compared with 41 (27%) women living in deprived areas (χ2=4.44, df=1, P=0.035).
Table 4.
Number of general practice consultations before and after diagnosis of breast cancer for women living in affluent and deprived areas
| Affluent
|
Deprived
|
Mann-Whitney test | ||||
|---|---|---|---|---|---|---|
| No of women | Median (interquartile range) | No of women | Median (interquartile range) | |||
| Year before diagnosis | 107 | 4 (1-7) | 160 | 5 (2-8) | Z=−1.88, P=0.06 | |
| 1st year after diagnosis | 108 | 10 (5-13) | 162 | 11 (6-16) | Z=−2.03, P=0.04 | |
| 2nd year after diagnosis | 107 | 5 (2-10) | 152 | 7 (4-13) | Z=−2.49, P=0.01 | |
In the first two years after the end of initial treatment, there was no difference in hospital admissions for problems related to breast cancer between women from affluent and deprived areas (number admitted once or more: 13 (10%) v 34 (16%), χ2 =2.62, df=1, P=0.106). However, women from affluent areas were less likely to be admitted than women from deprived areas (13 (10%) v 51 (24%), χ2 =11.1, df=1, P=0.001) for problems unrelated to breast cancer.
Discussion
We found no difference in access to hospital care or in surgical and non-surgical management of women from affluent and deprived areas with breast cancer, but women in deprived areas had more hospital admissions with problems unrelated to breast cancer and a higher general practitioner consultation rate, indicating that these women had a greater degree of comorbidity.
Our sample size was limited as we included only women under 75 years who lived in deprivation categories 1, 2, 6, or 7 and who had breast cancer diagnosed in 1992 or 1993. The decision to limit numbers in this way was pragmatic. It allowed detailed data from hospital (general and oncology) and general practice records to be collected by one researcher (UM), thus ensuring uniformity, and reduced the possibility of bias from changes in management.
Collecting data from general practice records was problematic as record keeping is not consistent between practices. Some practices were reluctant to participate, and we were able to examine only 86% of general practice records. Although the study was retrospective, the retrieval of hospital records was excellent, even for deceased patients.
Access to care
The largest component of delay between noticing symptoms and attendance at hospital clinic has been suggested to be patient delay.14 Although we are unable to comment on any delay before consultation in primary care, there was no significant delay in referral after presentation to general practitioners. The significant difference in the number of days between referral and clinic visit probably occurred because the hospital where many of the patients from affluent areas were seen had two clinics for new patients each week, whereas the other four hospitals had one clinic a week. In any case, the difference was not clinically important and did not cause a delay in surgery.
The numbers of days spent in hospital at the time of surgery were also similar. A study in south east England found that more women from deprived areas are admitted as emergencies and that more women from affluent areas are admitted as day cases.15 Information on emergency admissions is not available from our study, but the women from affluent areas did not have a shorter stay than those from deprived areas. Our data do not support the suggestion by Pollock and Vickers that “primary care is failing patients from deprived areas.”15
Treatment
We found no difference in the hospital management of women living in affluent areas and deprived areas in terms of type of breast surgery performed or whether they received radiotherapy, chemotherapy, or endocrine therapy. These results suggest that socioeconomic status was not a factor in these decisions.
We did, however, find a difference between the two groups with respect to axillary surgery. Closer inspection of the data shows that although axillary clearance was the operation of choice throughout the city, in one hospital almost half of axillary procedures were described as sampling (clearance 31 (52%), sampling 29 (48%)). This hospital saw women only from deprived areas, thus confounding the results. Since this study the variation has been eliminated, with more than 90% of women with breast cancer having axillary clearance (H Burns, personal communication, 1999) in keeping with recommendations in several recent clinical guidelines.16–18
Follow up
We found no differences between women living in affluent and deprived areas in terms of the total number of times seen at clinic. Although the numbers are small, more patients from deprived areas seemed to fail to attend appointments. This may relate to ease of access to hospital rather than lack of concern about follow up.
Analysis of the number of hospital admissions in the two years after diagnosis of breast cancer showed that admissions to hospital for conditions unconnected with breast cancer were more common in women living in deprived areas than among women from affluent areas. Consultations in primary care in the two years after diagnosis were greater for all women than before diagnosis, but were greatest for women living in deprived areas. Both the greater admission rate to hospital for conditions unrelated to breast cancer and the higher consultation rate in general practice may be an indication of greater comorbidity in women from deprived areas. This may explain the poorer survival figures that have been reported for all causes.19
Conclusion
The only differences in the management of women from affluent and deprived areas identified in this study could be explained by differences in hospital policy or were unlikely to have any significant impact on outcome. The NHS (in Glasgow, at least) seems to deliver health care equitably to women with breast cancer. The poorer survival of women from deprived areas with breast cancer may not be due to their breast cancer or its management but to other factors which result in deprived women (and also men) having a reduced life expectancy compared with affluent groups.19 In particular, the relation of comorbidity and deprivation requires further investigation.
What is already known on this topic
Breast cancer is commoner in affluent than deprived women
Survival, however, is greater among women from affluent areas than those from deprived areas
This is not explained by pathological stage at presentation
What this study adds
Access to hospital care for breast cancer is not related to social status
The NHS delivers health care equitably to women with breast cancer who live in affluent and deprived areas
Poorer survival of women living in deprived areas may be related to the presence of more comorbidity in deprived areas
Acknowledgments
We thank the West of Scotland Cancer Surveillance Unit, the medical records officers at the hospitals, and the general practitioners and practice staff for their help with this study. We thank the other members of the steering group for their input: Professor L Fallowfield, Professor J McEwen, and Dr Yvonne Taylor.
Footnotes
Funding: Cancer Research Campaign and National Lottery Charities Board (Scotland).
Competing interests: None declared.
References
- 1.Tomatis L, editor. Cancer: causes, occurrence and control. Lyons: International Agency for Research on Cancer; 1990. . (IARC Scientific Publication No 100.) [Google Scholar]
- 2.Gordon NH, Crowe JP, Brumberg DJ, Berger NA. Socioeconomic factors and race in breast cancer recurrence and survival. Am J Epidemiol. 1992;135:609–618. doi: 10.1093/oxfordjournals.aje.a116340. [DOI] [PubMed] [Google Scholar]
- 3.Karjalainen S, Pukkala E. Social class as a prognostic factor in breast cancer survival. Cancer. 1990;66:819–826. doi: 10.1002/1097-0142(19900815)66:4<819::aid-cncr2820660437>3.0.co;2-e. [DOI] [PubMed] [Google Scholar]
- 4.Vagero D, Persson G. Cancer survival and social class in Sweden. J Epidemiol Community Health. 1987;41:204–209. doi: 10.1136/jech.41.3.204. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Carnon AG, Ssemwogerere A, Lamont DW, Hole DJ, Mallon EA, George WD, et al. Relation between socioeconomic deprivation and pathological prognostic factors in women with breast cancer. BMJ. 1994;309:1054–1057. doi: 10.1136/bmj.309.6961.1054. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Schrijvers CTM, Mackenbach JP, Lutz JM, Quinn MJ, Coleman MP. Deprivation and survival from breast cancer. Br J Cancer. 1995;72:738–743. doi: 10.1038/bjc.1995.403. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Macleod U, Ross S, Gillis C, McConnachie A, Twelves C, Watt GCM. Socioeconomic deprivation and stage of disease at presentation in women with breast cancer. Ann Oncol. 2000;11:105–107. doi: 10.1023/a:1008385321476. [DOI] [PubMed] [Google Scholar]
- 8.Basnett I, Gill M, Tobias JS. Variations in breast cancer management between a teaching and a non-teaching district. Eur J Cancer. 1992;28A(12):1945–1950. doi: 10.1016/0959-8049(92)90233-r. [DOI] [PubMed] [Google Scholar]
- 9.Richards MA, Wolfe CDA, Tilling K, Barton J, Bourne HM, Gregory WM. Variations in the management and survival of women under 50 years with breast cancer in the South East Thames region. Br J Cancer. 1996;73:751–757. doi: 10.1038/bjc.1996.131. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Sainsbury R, Rider L, Smith A, MacAdam A. Does it matter where you live? Treatment variation for breast cancer in Yorkshire. Br J Cancer. 1995;71:1275–1278. doi: 10.1038/bjc.1995.246. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Sainsbury R, Haward B, Rider L, Johnston C, Round C. Influence of clinician workload and patterns of treatment on survival from breast cancer. Lancet. 1995;345:1265–1270. doi: 10.1016/s0140-6736(95)90924-9. [DOI] [PubMed] [Google Scholar]
- 12.Gillis CR, Hole DJ. Survival outcome of care by specialist surgeons in breast cancer. A study of 3786 patients (in a geographically defined area of 1.5 million population) in the west of Scotland. BMJ. 1996;312:145–148. doi: 10.1136/bmj.312.7024.145. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Carstairs V, Morris R. Deprivation and health in Scotland. Aberdeen: Aberdeen University Press; 1988. [Google Scholar]
- 14.Nichols S, Waters WE, Fraser JD, Wheeller MJ, Ingham SK. Delay in the presentation of breast symptoms for consultant investigation. Community Med. 1981;3:217–225. doi: 10.1007/BF02549119. [DOI] [PubMed] [Google Scholar]
- 15.Pollock AM, Vickers N. Deprivation and emergency admissions for cancers of colorectum, lung, and breast in south east England: ecological study. BMJ. 1998;317:245–252. doi: 10.1136/bmj.317.7153.245. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.British Breast Group. The provision of breast services in the UK: the advantages of specialist breast units. Edinburgh: Churchill Livingstone; 1995. [Google Scholar]
- 17.Breast Surgeons Group of the British Association of Surgical Oncology. Guidelines for surgeons in the management of symptomatic breast disease in the United Kingdom. Eur J Surg Oncol. 1995;21(suppl A):1–13. [PubMed] [Google Scholar]
- 18.Scottish Intercollegiate Guidelines Network. Breast cancer in women. Edinburgh: SIGN; 1998. (SIGN publication No 29). [Google Scholar]
- 19.Eames M, Ben-Shlomo Y, Marmot M. Social deprivation and premature mortality; regional comparison across England. BMJ. 1993;307:1097–1102. doi: 10.1136/bmj.307.6912.1097. [DOI] [PMC free article] [PubMed] [Google Scholar]